Background: For many years the serious mental illness of schizophrenia was thought of as the consequence of abnormalities in development which emerged in the course of adolescent brain development, but whose brain abnormalities did not change after onset. Now, however a new paradigm of post-onset progression of loss of brain gray matter and concomitant increase of brain fluid spaces (CSF) visible on structural MRI scans has become a hot topic in schizophrenia research. As well as other studies, our current Merit award-sponsored longitudinal MRI study and its 68 publications provide strong preliminary evidence of progression. Still, many questions remain. Not all investigators agree on progression of gray matter loss as a part of schizophrenia, and the role of treatment medication in progressive brain changes is controversial. Research Methods: To answer key questions about progression, we propose a 5 year longitudinal naturalistic study in two groups, first episode schizophrenia (FESZ) and chronic schizophrenia (CSZ), both compared with equal numbers of healthy controls (HC) matched on age, gender and parental socio- economic status who will have the same longitudinal protocol; all subjects will have 3T MRI scans. FESZ will have MRIs and clinical/diagnostic evaluations: at 1) onset, objectively defined as first hospitalization (N=110 total over 5 years); 2) 6 months after initial scan (N=63 total); 3) 12 months after initial scan (N=40 total); and 4) 30 months after initial scan (N=20 total). The FESZ subject numbers at each time point reflect both our documented follow-up return rate and the duration constraints of a 5 year study and are based on 22 new FESZ entering each year of the study. We will use the gold standard of manual Region of Interest analysis of MRI scans together with clinical symptom/sign measures to provide answers to key questions. To facilitate comparison FESZ and CSZ will receive the same initial and follow-up clinical and neuropsychological evaluations. Medication status will be tracked monthly by the treating physician and compliance, dosage and load will be tracked for association with MRI/clinical changes. An important feature of this naturalistic study is that power calculations based on preliminary data indicate the FESZ sample N will be large enough to permit subgrouping on the basis of medication, gender and other variables. We will use three measures of onset, the schizophrenia onset scale, medication onset and first hospitalization and will determine which provides the best measure of onset, based on predictive value for MRI changes. Hypotheses: We hypothesize, based on preliminary data, that post-onset progression is most rapid in FESZ in the first 6 months after onset, compared with 28 CSZ at initial and at 30 month follow-up. In FESZ we expect gray matter progressive loss in widespread neocortical areas, most intense in frontal and temporal lobe, and, within these lobes, most intense and rapid in particular brain regions linked to SZ functional abnormalities, such as, for example, in superior temporal gyrus language and auditory processing regions, where it will be associated with progression of clinical symptoms of disorganization and thought disorder. We expect antipsychotic medication to slow progression of both gray matter loss & clinical symptoms while mood stabilizers will slow gray matter progression but will have lesser effect on clinical symptoms.